Browsing by Author "Yilmaz,A.E."
Now showing 1 - 2 of 2
- Results Per Page
- Sort Options
Article Citation Count: 0Can computers read a text with stress?;(2013) Uslu, İbrahim Baran; Demir,N.; Ilk,H.G.; Yilmaz,A.E.; Department of Electrical & Electronics EngineeringToday prosodic elements such as stress, intonation and melody can be examined through computer-assisted techniques. The identification and analysis of acoustic qualities provide significant clues as to the suprasegmental aspects of speech. In this study a system was designed to enable speech synthesis using a Turkish text. This synthesizer was used to analyze the melodic sructures of selected sentences. Any speech synthesized by a computer needs to be equipped with a melodic model so that it will sound natural to the ear. Within the context of this study a melodic model was also suggested, and the elements of this model were analyzed and discussed in terms of the rules of linguistics.Article Citation Count: 0A rule based prosody model for Turkish text-to-speech synthesis;(2013) Uslu, İbrahim Baran; Ilk,H.G.; Yilmaz,A.E.; Department of Electrical & Electronics EngineeringThis paper presents our novel prosody model in a Turkish text-to-speech synthesis (TTS) system. After developing a TTS system driven by parametric features consisting of duration, pitch and energy modifications, we try to figure out some prosody rules in order to increase the naturalness of our synthesizer. Since the inflected verbs in Turkish can be stand-alone sentences with the suffixes they take, we build a perceptual prosody model by defining rules on the stress patterns of verb inflections. Affirmative, negative and interrogative (both positive and negative) forms of many verbs were examined in a systematic way. Not only verbs, but in the same way, some phrases were examined for obtaining a proper prosody. According to the results of listening tests, the defined rules based on duration, pitch and energy modification weights, result in perceptually better speech synthesis, namely about 1,78/5,0 improvement in average in the CMOS (Comparative Mean Opinion Score) test. This improvement shows the success of our novel prosody model.